番茄社区

Debate over butter flared this past year, with conflicting reports on whether the stuff is unhealthy. In this exclusive 番茄社区 video, , of Harvard T.H. Chan School of Public Health in Boston, addresses the surrounding confusion, emphasizing the key issue of comparators. While occasional butter consumption is permissible, he argues that there are healthier fats (such as olive and soybean oils) that improve patients' chances for maintaining long-term health.

Following is a transcript of his remarks:

There has been a lot of confusion about the role of butter in a diet, and probably the most important reason for this confusion has been the failure to specify what butter is being compared with. If you compare butter with the old high trans margarines, almost for sure butter was the healthier choice, but those old margarines are pretty much off the market now and almost all of them are trans free at this point in time.

If you compare butter with calories from refined starch and sugar, it's going to be pretty much a wash. They'll both have adverse impacts on metabolic factors and on risk of heart disease and diabetes. However, if you compare butter with the liquid plant oils like soybean oil, olive oil, canola oil, pretty much all the liquid vegetable oils, those plant oils are going to be a whole lot better than butter.

What should a physician be conveying to their patients? It's basically this information: whenever you have a chance to replace butter with something that's healthier, basically the liquid vegetable oils, that's going to be a better choice, both for improving blood lipids and for reducing risk of heart disease in type 2 diabetes.

There have been some studies that have been fueling some of the confusion about butter. For example, there was a very prominent meta-analysis in the Annals of Internal Medicine about a year and a half ago which seemed to conclude that the type of fat in the diet didn't make any difference with regard to heart disease.

However, that was an extremely flawed meta-analysis in multiple ways. One thing, there were just egregious errors in pulling the data out of the original studies, but most fundamentally they actually couldn't do the right analysis because they were trying to summarize papers in which the comparison for saturated fat wasn't specified, and it's really important whether you're comparing saturated fat with starch or with trans fat or with healthy plant oils that are mostly unsaturated, that makes a huge difference, and if you don't specify the comparison, most of the calories in our diet are coming from refined starch, sugar, and red meat.

Therefore, if you're comparing saturated fat with all those really unhealthy calories, saturated fat is not going to look so bad. It's going to look about the same. But if you compare saturated fat with healthy plant oils, using those healthy plant oils will definitely reduce the risk of heart disease while they're improving blood lipids at the same time.

As we've come to look at data that's been accumulating in our long-term studies after the last three or four decades, what we've seen is that the type of fat in the diet is extremely important. Total fat percentage of calories and total fat just doesn't seem to make much difference. So we see huge benefits of replacing trans fat and saturated fat with unsaturated plant oils for heart disease. But in some of our more recent analyses, we've seen major benefit for total mortality as well, and it's not coming just from heart disease. It's coming from multiple other endpoints as well, including neurologic diseases, lower risk of neurologic diseases with healthier fats, lower risk of chronic lung disease and other conditions. So making healthy types of fat in the diet a priority is one of the most important things people can do about their long-term health and well-being.

We've tried to put together some of the most up-to-date studies as well as background literature on our Department of Nutrition website, Nutrition Source. Just go to Google, put in Nutrition Source, and you'll be there.

Comment